Rogowski coil calibration / Electromagnetic Projectile Accelerators / Forums

Forums

4hv.org :: Forums :: Electromagnetic Projectile Accelerators

« Previous topic | Next topic »

Rogowski coil calibration

Move Thread

LAN_403

TheMerovingian

Sun Oct 17 2010, 07:45PM

Registered Member #14 Joined: Thu Feb 02 2006, 01:04PM
Location: Prato/italy
Posts: 383

I have done some acquisitions of current waveforms using a rogowski coil with active integration (yet is has some light offset problems caused by insufficient slew rate).

The problem is that , while this system has precise timebase values, the vertical value is unknown. THe value of current shown is based on mutual inductance estimation of the rogowski coil. Is there a way to calibrate the system whithout using a massive one-turn transformer and a current clampmeter?

Show me your experiences with rogowski coils
]coilgun.pdf[/file]

klugesmith

Wed Oct 20 2010, 05:06AM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1714

Hi there. Nice pictures you got -- any idea what makes the extra bumps?

As for the thread title question: wish I'd had time to finish the thread started here:

(Now have eleven printed-circuit Rogowski coils, geometrically identical, and could send you one to play with).

As you know, the sensitivity of a Rogowski coil is its mutual inductance with a single loop carrying the unknown current. Common values are between 10 nH and 100 nH. I have measured M by two methods, with good agreement:

1. As a reference current source, use a lightly damped RLC discharge. If you know the capacitance, measure the frequency of ringing to determine L. Then, knowing the initial DC voltage, you can calculate the current amplitude. In my case, about 50 uF was charged with a 9V battery. 15 kHz ringing implied L = 2 uH and peak current of about 40 amperes.
Your reference current source could be a real coilgun circuit, especially if linear (nonpolarized capacitor, no clamp diode, mechanical switch) so it's easy to determine the RLC parameters from just C, measured frequency, and measured damping ratio.

2. The measured self-inductance of a Rogowski coil, divided by the total number of turns will give the mutual inductance or Rogowski sensitivity. Self-inductance value must exclude L of the cable to integrator or scope. In my case about 2.1 uH (that's a coincidence) divided by 60 turns gives about 35 nH.

p.s. if you can get digital storage oscilloscope traces into your computer, no need for analog integrator. Capture the raw Rogowski voltage, then integrate on the computer. 50 ohm termination at the scope helps to damp high frequency (many MHz) ringing in the instrument circuit, when the main coilgun switch closes and di/dt "instantly" jumps from zero to its maximum value. Rogowski voltage source is low impedance so the 50 ohms won't significantly load it down.

Steve Conner

Wed Oct 20 2010, 08:58AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Klugesmith wrote ...

Rogowski voltage source is low impedance so the 50 ohms won't significantly load it down.

Are you sure? I thought the output impedance of a Rogowski coil was its self-inductance, and surely that can be significant at high values of di/dt.

I think if you run one into a low enough impedance, it operates in self-integrating mode with a time constant of L/R.

I used Rogowski coils to measure primary current in my OLTC work. I was running primary currents over 5kA peak, which was a bit too much for a ferrite-cored CT. I always "calibrated" them by calculating the sensitivity from theory and assuming it to be correct

The results agreed well with what I predicted, given the tank capacitor charging voltage and the surge impedance of the tank circuit. In fact, I once half seriously recommended using an OLTC primary circuit as a Rogowski coil calibrator.

klugesmith

Wed Oct 20 2010, 12:04PM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1714

Great minds think alike, eh?

We both made Rogowski-able systems that are nice and linear,
with low damping, whose current can be calculated from voltage, capacitance, and frequency.

We agree that the RC output impedance is its self-inductance
(below self resonant frequency, which is over 20 MHz in my case).

I think that could be significant with a RC wound on coax, at your TC frequency,
and not with my 60-turn RC at can-crusher frequency (15 kHz)
or coilgun frequency (few 100 Hz).
Self-inductance of 2 uH works out to 0.2 ohms at 15 kHz.
Conversely, the corner frequency for roll-off due to Lself / Rload
is almost 4 MHz.

When I first saw the strong ringing on unintegrated initial step (di/dt rises to max in the time it takes for LCR discharge switch to close), I thought it was reflections in the unterminated cable between RC and scope. Now I think it's more likely the RC self-resonance going.

Dr. Shark

Tue Dec 07 2010, 05:00AM

Registered Member #75 Joined: Thu Feb 09 2006, 09:30AM
Location: Montana, USA
Posts: 711

A third vote for the "use an RCL resonator with known characteristics" method here. I had a Rogowski coil in a disk launcher with know capacitance and inductance and just simulated the whole mess, tweaking parameters for stray resistance and inductance until the simulation agreed with the measured current and voltage waveforms. Then you can simply read off the current from the simulation and use that to determine the sensitivity of the coil.

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob